Re: Separate Data and Log files

Tech-Archive recommends: Repair Windows Errors & Optimize Windows Performance

From: Andrew J. Kelly (sqlmvpnooospam_at_shadhawk.com)
Date: 01/06/05


Date: Thu, 6 Jan 2005 13:37:34 -0500

Those numbers don't really mean much by themselves other than the cachehit
ration. If it is only 80% you are most likely either way low on memory or
simply have a lot of untuned queries. MS does not have a preferred SAN
vendor per say although if you are going to cluster it you need to ensure it
is on the Clustering HCL. Most of the modern SAN vendors (EMC, Hitachi,
Fugitsu etc.) will work fine given a proper configuration. I am not sure
what type of document you are looking for but usually the configuration of a
SAN on a higher end system requires a fair amount of coordination with the
SAN vendor.

-- 
Andrew J. Kelly  SQL MVP
"Tom W" <TomW@discussions.microsoft.com> wrote in message 
news:F0CA38B9-5D31-4454-9B5D-B628690EBBAC@microsoft.com...
> The servers are typically running about 400 transaction/sec, about an 80%
> data cache hit and cpu utilization of aroun 40%.   There is moderated 
> writes.
> Does Microsoft have anything documented or maybe even some SAN vendor?
>
>
>
> "Andrew J. Kelly" wrote:
>
>> It is quite possible that a SAN can mask the need to separate the files 
>> due
>> to large amounts of cache. But that still depends on what you are doing, 
>> how
>> many transactions etc. I have seen SAN's in which the files were not
>> segregated and there were performance issues but it all depends.
>>
>>
>> -- 
>> Andrew J. Kelly  SQL MVP
>>
>>
>> "Tom W" <Tom W@discussions.microsoft.com> wrote in message
>> news:DE458441-263C-43AF-9A17-E8C719283AD9@microsoft.com...
>> > Is there a documented information about separating the data and log 
>> > files
>> > pertaining to physical disk layouts.  How much relevance is this
>> > documentation with newer SAN technologies where there are large buffer
>> > caches.  Because my understanding is that the SAN units notifies the
>> > server
>> > that the data is written to disk when in actuality it is still in cache
>> > waiting to be written to disk.
>> >
>> > Tom
>>
>>
>> 


Relevant Pages

  • Re: lseek/write
    ... opposed to not using lseek. ... in cache and with it, ... When I call fsync i flush the cache to disk but only the new ... There's really no reason they should be, depending on how your SAN is ...
    (comp.os.linux.development.system)
  • Re: Exchange 2007 SCC Cluster - Slow Failover
    ... Doesn't surprise me a bit. ... A SAN is not a panacea for performance issues, ... the SAN and the disk layout? ... cache to a copy of the database that won't be used". ...
    (microsoft.public.exchange.admin)
  • RE: Write Caching
    ... I have seen FTDISK events being reported even for SAN volumes. ... In other words the cache between ... > the NODES Raid Controller cards has no way to update. ... >> PERC 4DC are in cluster mode and by default set to write through caching. ...
    (microsoft.public.windows.server.clustering)
  • San Performace with SQl Server
    ... Am presenlty Testing the I/o performance on the new CX-500 San From Emc ... And Hba From Emulex,The LUn Size is 1 TB and when doing the Testing ... requests i get 6.04 MB/sec as the output,the read/Write cache is ...
    (microsoft.public.sqlserver.clustering)
  • Re: Backup / Restore Estimates
    ... If I understand the above correctly, you have 8GB of onboard cache ... on the SAN controllers and you have 30GB of database to dump. ... we'll be bound by the speed of the disks. ... Pablo Sanchez - Blueoak Database Engineering, ...
    (comp.databases.sybase)